Innovative Food Science and Emerging Technologies 20 (2013) 152–160
Contents lists available at ScienceDirect
Innovative Food Science and Emerging Technologies
journal homepage: www.elsevier.com/locate/ifset
Effects of polar cosolvents on cocoa butter extraction using supercritical
carbon dioxide
E.K. Asep a,b, S. Jinap a,c,⁎, M.H.A. Jahurul d, I.S.M. Zaidul e, H. Singh f
a
Food Safety Research Centre (FOSREC), Faculty of Food Science and Technology, Universiti Putra Malaysia, 43400 UPM, Serdang, Selangor, Malaysia
Department of Food Technology, Faculty of Engineering, University of Pasundan, 40153 Bandung, West Java, Indonesia
Institute of Tropical Agriculture, Universiti Putra Malaysia, 43400 UPM, Serdang, Selangor, Malaysia
d
School of Industrial Technology, Universiti Sains Malaysia, Minden, 11800 Penang, Malaysia
e
Department of Pharmaceutical Technology, Kulliyyah of Pharmacy, International Islamic University Malaysia, Kuantan Campus, 25200 Kuantan, Pahang, Malaysia
f
Department of Chemical Engineering, Faculty of Engineering, Universiti Malaya, 50603 Kuala Lumpur, Malaysia
b
c
a r t i c l e
i n f o
Article history:
Received 16 June 2012
Accepted 28 June 2013
Editor Proof Receive Date 25 July 2013
Keywords:
Cocoa butter
Fatty acid
Polar cosolvent
Supercritical carbon dioxide
Selectivity
Triglycerides
a b s t r a c t
Cocoa butter was successfully extracted from cocoa liquor by supercritical carbon dioxide (SC–CO2) at 35 MPa,
60 °C and 2 mL/min with 5%, 15% and 25% cosolvents. The extraction yield of triglycerides (TG) and fatty acid
(FA) compositions were significantly influenced by the concentration of polar cosolvents. The SC–CO2 extraction
efficiency was increased with cosolvent significantly. Ethanol was found to be the best cosolvent for cocoa
butter extraction using SC–CO2 followed by isopropanol and acetone. The triglycerides of 1,3-dipalmitoyl-2oleoylglycerol (POP), 1-palmitoyl-2-oleoyl-3-stearoyl-glycerol (POS) and 1,3-distearoyl-2-oleoyl-glycerol
(SOS) were contained in the extracted cocoa butter with POS being the major component. Where palmitic,
stearic and oleic were the main fatty acids in the cocoa butter samples, with stearic being the highest component.
The lower molecular weight (MW) of TGs and FAs showed the higher selectivity compared to the high MW of
TGs and FAs. Thus, they were fractionated during the first stage of SC–CO2 process.
Industrial relevance: The cocoa butter was successfully extracted from cocoa liquor by SC–CO2 at 35 MPa, 60 °C
and 2 mL/min using different concentrations of polar cosolvents (ethanol, isopropanol and acetone). The extraction yield was significantly (p b 0.05) influenced by the concentration of polar cosolvents. Similarly, polar
cosolvent concentration had significant (p b 0.05) effects on the TG and FA compositions. Ethanol was found
to be the most efficient polar cosolvent for cocoa butter extraction compared to isopropanol and acetone. POS
(42.2–45.9%) being the major triglycerdies component, followed by SOS (27.6–31.4%) and POP (20.3–22.7).
Palmitic, stearic and oleic acids were the main fatty acids in the extracted cocoa butter, with stearic being the
highest (34.9-37.8%), followed by oleic (30.3-31.8%) and palmitic (28.3-30.0%) acids, respectively. The choice
of modifiers becomes a great challenge and ethanol was shown to be the best polar cosolvent, and it enhanced
the solubility during the cocoa butter extraction by SC–CO2. This method can be feasibly implemented in the
cocoa industry for the production of high quality cocoa butter.
© 2013 Elsevier Ltd. All rights reserved.
1. Introduction
Cocoa butter is highly demanded by food, cosmetic and pharmaceutical industries. Mechanical expression and solvent extraction
with hexane are generally employed for obtaining cocoa butter. However, there is an increasing concern of the health and safety hazards
associated with the use of organic solvents, while expression by hydraulic method often introduces contaminants into the cocoa butter
that must be removed later. Greater concern over the disposal of
such toxic organic solvents and their effect on the environment has
⁎ Corresponding author at: Food Safety Research Centre (FOSREC), Faculty of Food
Science and Technology, Universiti Putra Malaysia, 43400 UPM, Serdang, Selangor,
Malaysia. Tel.: +60 3 89468393; fax: +60 3 89423552.
E-mail addresses: [email protected], [email protected] (S. Jinap).
1466-8564/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ifset.2013.06.010
led to a move towards cleaner extraction methods such as supercritical fluid extraction (SFE). Supercritical fluid extraction, mainly by supercritical carbon dioxide (SC–CO2) is a potential alternative to the
customary methods of producing cocoa butter. It offers the advantages of rapid, nontoxic, environmental-friendly, contamination-free
and easily manipulated conditions (Li & Hartland, 1996).
The efficiency of SFE process mainly depends on the solvent capacity of supercritical carbon dioxide. This can easily be modified by
varying pressure and temperature (Dauksas, Venskutonis, & Sivik,
1998; Francisco & Dey, 2003; Reverchon & Camillis, 1991). Due to
the poor solvent capacity of supercritical CO2, the addition of small
quantity of organic solvent as a cosolvent or modifier or entrainer
has also been recommended to increase the solubility of the analyte
or possibly to increase the separation of coextractives (Dobbs,
Wong, Lahiere, & Johnston, 1987; King, 1993a,b; King & France,
E.K. Asep et al. / Innovative Food Science and Emerging Technologies 20 (2013) 152–160
1992; Reverchon & Camillis, 1991; Stahl, Quirin, & Gerald, 1988). In
fact, the organic solvent can improve selectivity of the target analytes
extracted by SC–CO2 from the sample matrix, thus resulting in higher
extraction efficiency. Performing SFE procedure along with cosolvents
usually results in higher extraction efficiency than that obtained by
pure CO2.
As mentioned earlier, an alternative for improving the solvent capacity of nonpolar carbon dioxide is to use another solvent with higher polarity. However, there are several practical limitations. Supercritical
ammonia would be very attractive from the view point of solvent
strength, but it is difficult to pump since it tends to dissolve pump seal.
Furthermore, it is also a chemically reactive solvent, thus making it too
dangeraous for routine use (Hawthorne, 1990). Other solvents like
methanol and isopropanol may also be exellent supercritical solvents,
but their high critical temperatures make it difficult to attain the supercritical states. Other cosolvents such as acetone and isopropanol have
also been used as the second polar cosolvent for commercial deoiling
processes. Hence, a polar modifier is usually added to supercritical
carbon dioxide to remove the highly polar organic compounds. The modifier enhances the solubility of SC–CO2, thereby increasing the extraction
efficiency. Moreover, the modifier interacts with the analyte/matrix
complex to promote rapid desorption into the supercritical fluid
(Luque de Castro & Garcia-Ayuso, 1998).
Ethanol is commonly used as cosolvent or modifier for the extraction
of natural products because the toxicity of ethanol to the human body
is low (Catchpole, Grey, & Noermark, 1998; Temelli, 1992; Walsh,
Greenfield, Ikonomou, & Donohue, 1989; Walsh & Ikonomou, 1987). Furthermore, it can be easily removed from the food matrix although there
is a limited information regarding the effect of ethanol and other
cosolvents on the SC–CO2 extraction of cocoa butter. Thus, the objective
of this study was to investigate the effect of type and concentration of
polar cosolvents (ethanol, isopropanol and acetone) on the extraction efficiency and selectivity, and TG and FA profiles of cocoa butter extracted
by using SC–CO2. The overall quality of extracted cocoa butter was determined by the qualitative and quantitative analyses of TG profile and FA
composition. The SC–CO2 extraction process was carried out at different
concentrations (5, 15 and 25%, mol%) of cosolvent.
2. Materials and methods
2.1. Materials
Cocoa liquor samples (size of diameter = ~ 0.074 mm) were purchased from K.L. Kepong Sdn. Bhd., Port Klang Malaysia. Liquid CO2
with a purity of 99.9% was obtained from Malaysian Oxygen (MOX),
Petaling Jaya, Selangor, Malaysia. Chromatography grade solvents
(i.e. petroleum ether, methanol and acetonitrile) and cosolvents (i.e.
ethanol, acetone, isopropanol) and standard of triglycerides, fatty
acid methyl esters were supplied by Sigma Aldrich Sdn. Bhd. (Petaling
Jaya, Malaysia), Chemolab Sdn. Bhd. (Kuala Lumpur, Malaysia) and
Fisher Scientific Sdn. Bhd (Shah Alam, Malaysia), respectively.
2.2. Supercritical fluid extraction (SFE) method
The SFE apparatus consisted of 2 intelligent high performance liquid chromatography (HPLC) pump (model PU-1580, Jasco Corporation, Tokyo, Japan), i.e. CO2 pump and cosolvent pump. The CO2
pump was fitted with a cooling jacket to deliver CO2 and cosolvent
pump to deliver cosolvents (as modifier/entrainer). In order to cool
the pump head of CO2 pump, ethylene glycol-deionized water mixture (50:50, v/v) was circulated through the cooling jacket using a
low temperature bath circulator (model 631D, Tech-Lab Manufacturing Sdn. Bhd., Petaling Jaya, Malaysia) which can deliver coolant
down to − 20 °C. A 10 g of sample was loaded into a 50 mL extraction
vessel (model EV-3, Jasco Corporation, Tokyo, Japan) that was placed
in column oven (model CO-1560, Jasco Corporation, Tokyo, Japan).
153
The column oven was used to maintain the extraction temperature.
Both CO2 and cosolvent pump were connected to extraction vessel
which was placed inside the column oven to maintain the extraction
flow rate of fluid. A back pressure regulator (BPR) (model BP-1580-81,
Jasco Corporation, Tokyo, Japan) was used to control the extraction pressure. The extraction was carried out at 35 MPa pressure, 60 °C extraction
temperature and 2 mL/min flow rate using the mixture of cosolvent
and carbon dioxide. Ethanol, isopropanol and acetone were used as
polar modifiers at the cosolvent/CO2 concentration ratios of 5, 15 and
25% (mol%). A 100 mL of blue cup bottle was used as a collection vessel.
The concentrations of cosolvent were determined based on the
actual volume of cosolvent delivered into a known volume of liquid
carbon dioxide. In this approach, the actual molar percentage of
cosolvent to carbon dioxide would change as the density of carbon
dioxide varied as the pressure changed. The molar percentage of
cosolvent in carbon dioxide was calculated by the following equation:
ðVcosolventÞðρcosolventÞ
MWcosolvent
mol% cosolvent in CO2 ¼ ðVcosolventÞðρcosolvent
Þ
ÞðρCO2Þ
þ ðVCO2
MWCO2
MWcosolvent
ð1Þ
where V is the volume, ρ is the density at 35 MPa and 60 °C, and MW
is the molecular weight.
2.3. Determination of yield
The initial weight of the 100 mL test tubes used for collecting the
yield of the extracted cocoa butter was measured gravimetrically
using 3 digital balance (Mettler-Toledo (M) Sdn Bhd, Selangor,
Malaysia). After SFE extraction, the test tubes containing the extracted
fat were transferred into a desiccator and held at room temperature
until the constant weight was obtained. The residue of cosolvents was
evaporated from the extracted fat using a rotary evaporator (model
WB/VV 2000, Heidolph, Schwabach, Germany) under vacuum at
temperature of 70 °C; the extracted fat was then placed in an oven
(model ULM 500, Memmert, Schwabach, Germany) at 45 °C for
30 min and later transferred into the desiccator for drying. Yield was
calculated by dividing the percentage of cocoa butter in the samples
with 53% factor amount of cocoa butter in cocoa nibs, determined by a
Soxhlet method using petroleum ether accoding to AOAC (AOAC,
1998). The expression of yield was calculated by the following equation:
h
Yield ðwt% fractionÞ ¼
weight of fat
weight of sample
i
53%
100%
:
ð2Þ
2.4. Determination of triglycerides
Triglycerides composition was determined by HPLC according to the
AOCS method (AOCS, 1993). Ten percent cocoa butter solution was prepared using acetone as solvent. The solution was then filtered using TE
36 membrane filter (PTFE; 0.45 μm) (Millipore, Massachusetts, USA)
before being filtered through the Sep-Pak Plus Silica cartridge (Waters,
Virginia, USA) to discard any impurities. Determination of triglyceride
was conducted using Waters (Virginia, USA) HPLC instrument, model
600 controller and model 410 RI detector, and C18 column (3.9 mm
id × 300 mm length), with a column temperature of 30–35 °C, column
pressure of 5–6 MPa, flow rate of mobile phase (acetone/acetonitrile
75:25 v/v) of 1 mL/min and injection volume of 10 μL. The value of triglycerides was expressed as a percentage. All analyses were done in
three replications and duplicate injections.
2.5. Determination of fatty acid methyl esters (FAMEs)
The fatty acid composition of fat mixtures was determined as fatty
acid methyl esters (FAMEs) using gas chromatography (AOCS, 1993).
Restek Rtx-2330 column (Restek Corporation, Pennsylvania, USA)
E.K. Asep et al. / Innovative Food Science and Emerging Technologies 20 (2013) 152–160
(30 m; 0.25 mm; 0.2 μm) was used. FAMEs were prepared by dissolving 0.05 g sample into 0.8 mL petroleum ether (b.p. 40–60 °C) and
0.2 mL of 1 M sodium methoxide (30% methanol in sodium
methoxide) was added. The mixture was then shaken gently using
an autovortex (Stuart, Manchester, U.K.) for 30 s and kept for 5 min
for it to form two layers. A volume of 0.5 μl of the upper layer was
injected into an injector port of gas chromatography GC HP 5890A
(Hewlett Packard, Wilmington, USA) with the temperature programme that started at 115 °C, a heating rate of 6 °C/min and a final
temperature of 200 °C (AOCS, 1993). Analysis was carried out at 5,
10 and 15 h extraction time of cocoa butter with three replications.
2.6. Selectivity (α)
110
100
90
80
70
Yield (wt%)
154
50
40
The effect of cosolvents on TG profile and fatty acid (FA) composition of cocoa butter extracted by SC–CO2 process was determined by
estimating the selectivity (α) using the following equation:
ðX Þ = ðX Þ
ðα Þ ¼ 1 Ex 1 S ðX 2 ÞEx = ðX 2 ÞS
60
30
20
10
ð3Þ
0
0
where (X1)Ex and (X2)Ex denote concentration of two different TGs or
FAs in the extract phase (supercritical fluid); while (X1)S and (X2)S are
the concentration of two different TGs or FAs in the solid phase
(Soxhlet extraction) (Li & Hartland, 1992; Rossi, 1996; Rossi,
Arnoldi, Salvioni, & Schiraldi, 1989; Scheider, Kautz, & Tuma, 2000).
The selectivity (α) of TG was calculated by comparison of TG relative
to POS (C52) as the major TG of cocoa butter. Similarly, the selectivity
(α) of FA was calculated by comparison of FA relative to stearic acid
(C18:0) as the major FA of cocoa butter. The cocoa butter was also
extracted by a Soxhlet method (AOAC, 1998).
2.7. Statistical analysis
The data obtained from the measurements were subjected to analysis of variance (ANOVA) to determine the significant differences
among all treatments using the SAS software Version 8 (TS M1, SAS
Institute Inc., NC, USA). The experimental data were reported as the
mean ± SD of independent trials. Significant differences among all
treatments were determined by Duncan's multiple test at significant
level of p b 0.05.
2
4
6
8
10
12
14
16
18
20
22
24
Extraction Time (h)
5% Ethanol
15% Ethanol
25% Ethanol
Fig. 1. The yield of cocoa butter extracted from cocoa liquor using ethanol as cosolvent
in SC–CO2 at 35 MPa, 60 °C and 2 mL/min as a function of ethanol concentration and
the extraction time.
increase the solubility of cocoa butter. Li and Hartland (1996) reported that the addition of ethanol as a cosolvent significantly (p b 0.05)
enhanced the solubility of cocoa butter depending on the concentration of ethanol, thereby increasing the extraction efficiency of cocoa
butter up to 25–33% w/w of ethanol where the extraction efficiency
became constant. Cocero and Calvo (1996) reported that the solubility of sunflower oil in CO2 is enhanced at a rate of 5 g/kg CO2 for each
percent (by weight) of ethanol added to SC–CO2 at 30 MPa and 42 °C.
The ability to dissolve cocoa butter by using a sufficient amount of
100
90
3. Results and discussion
80
3.1. Effects on the yield
Yield (wt%)
70
The effects of ethanol as a polar cosolvent on the efficiency of
cocoa butter extraction using SC–CO2 are shown in Fig. 1. The results
show that the concentration of ethanol had a significant (p b 0.05) effect on the extraction yield of cocoa butter extracted by SC–CO2 at
35 MPa, 60 °C and a flow rate of 2 mL/min. The extraction yield increased as the concentration of ethanol was increased. The addition
of 25% ethanol (E3) resulted in 100% extraction efficiency after16 h
of extraction time; the efficiency decreased to 97.50% and 92.45%
with the addition of 15% and 5% ethanol at the same extraction
time. Asep, Jinap, Russly, Harcharan, and Nazimah (2008) reported
that a supercritical fluid of cocoa butter extraction from cocoa liquor
using SCF with pure CO2 for nearly 16 h and found a relatively high
extraction yield of 83.25%, and prolonging the extraction time up to
28 h resulted in 100% extraction yield. This indicates that the yield
of cocoa butter extracted increases with cosolvent concentration.
Every concentration of ethanol used in the study produced a higher
yield of cocoa butter compared to that of pure CO2 and significantly
(p b 0.05) increased the SC–CO2 extraction efficiency. The hydroxyl
group in the chemical structure of ethanol may form a hydrogen
bond to enhance the solute solubility. Ethanol may also substantially
60
50
40
30
20
10
0
0
2
4
6
8
10
12
14
16
18
Extraction Time (h)
5% Isopropanol
15% Isopropanol
25% Isopropanol
Fig. 2. The yield of cocoa butter extracted from cocoa liquor using isopropanol as
cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min as a function of isopropanol concentration and the extraction time.
E.K. Asep et al. / Innovative Food Science and Emerging Technologies 20 (2013) 152–160
100
(p b 0.05) effect on SC–CO2 extraction efficiency. The results indicate
that an increase in the concentration of isopropanol would result in
the increase of cocoa butter solubility. The SC–CO2 extraction using
25% isopropanol (I3) for 16 h resulted in the highest extraction yield
(96.70%) of cocoa butter; the yield decreased by 68.26% and 46.01%
when 15% and 5% isopropanol were added as a cosolvent, respectively. The extracted cocoa butter obtained by a low concentration (i.e.
15% or 5% w/w) of isopropanol was found to be less than that by
pure CO2 (Asep et al., 2008). The yield obtained from isopropanol
was similar to that from ethanol as the polarity of isopropanol
(1.66 D) was quite similar to that of ethanol (1.69 D).
The extraction yield of cocoa butter was significantly (p b 0.05)
influenced by the concentration of acetone (Fig. 3). The SC–CO2 extraction using 15% acetone (AC2) for 16 h resulted in the highest extraction yield (84.07%), while the smallest extraction efficiencies of
76.50% and 70.20% were achieved by the addition of 25% and 5% acetone, respectively. As also shown in Fig. 3, a relatively high extraction
yield (76.50%) was obtained by using 25% acetone (AC3). This finding
was consistent with those reported by Machmudah, Shotipruk, Goto,
Sasaki, and Hirose (2006) who reported that at higher concentrations
of acetone, low amounts of astaxanthin were extracted by SC–CO2
due to the decrease in the density of the supercritical fluid. Furthermore, the extraction selectivity of astaxanthin may also become
smaller, as other components of the feed matrix could be easily
extracted.
The nonpolar solvents tend to most efficiently dissolve nonpolar
solutes such as hydrocarbons, while more polar solvents tend to
most efficiently dissolve more polar solutes. The dipole moment of
acetone (2.91 Debye, D) was more polar than ethanol (1.69 D) and
isopropanol (1.66 D). However, because acetone had a smaller molecular interaction with the solute through hydrogen bounding, the
yield of cocoa butter extraction was much lower.
90
80
Yield (wt%)
70
60
50
40
30
20
10
0
0
2
4
6
8
10
12
14
16
18
Extraction Time (h)
5% Acetone
15% Acetone
155
25% Acetone
Fig. 3. The yield of cocoa butter extracted from cocoa liquor using acetone as cosolvent
in SC–CO2 at 35 MPa, 60 °C and 2 mL/min as a function of acetone concentration and
the extraction time.
ethanol as a cosolvent was found to be higher than that obtained by
either CO2 or ethanol alone. The cosolvent can facilitate selective separation of solutes with different polarities, hydrogen bonding and
abilities for association or complexation ability (Dobbs et al., 1987).
The solvent power of SC–CO2 could be increased by the addition of
a small amount of cosolvents, and the effect is dependent on the concentration of the cosolvent in the supercritical phase, which is determined by the phase behaviour of the mixture under operating
conditions (Olimpio et al., 2005). Shi et al. (2009) also reported that
the effect of ethanol as a polar cosolvent is higher than the effect of
water on lycopene yields at both temperatures because ethanol can
interact with the matrix and facilitate analyst adsorption.
Fig. 2. shows the effects of isopropanol concentration on the extraction yield of cocoa butter obtained at 35 MPa, 60 °C extraction
temperatures and a flow rate of 2 mL/min. A significant (p b 0.05)
difference was found between the extraction yields obtained at
different concentrations of isopropanol; the extraction yield significantly (p b 0.05) increased with an increase in the concentration of
isopropanol. In fact, the concentration of isopropanol had a significant
3.2. Effects on triacylglycerol profile and selectivity (α)
As shown in Tables 1 to 3, several TG profiles of cocoa butter
extracted by SC–CO2 at 35 MPa, 60 °C and 2 mL/min were significantly (p b 0.05) affected by the cosolvent (i.e. ethanol, isopropanol and
acetone) concentrations except in the effect of ethanol (I) on POS
(E2, E3), POP (5 h), POS (10, 15 h) and other TGs (5, 10 h), in the effect of isopropanol (I) on POS (I3), other TGs (I3), POP (5 h), POS
(5 h), SOS (5 h), and other TGs (5 h), and in the effect of acetone
(AC) on POP (AC3), other TGs (AC1) and others TG (10, 15 h). The results demonstrated that cocoa butter extracted under different processing conditions mostly contained three main TGs namely POS,
POP and SOS.
Table 1
Triacylglycerols composition (area %)α changes of cocoa butter extracted from cocoa liquor with ethanol as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min with different ethanol concentration and extraction time.
Sampleβ
Triacylglycerols compositions
5 h extraction time
POP
E1
(5%)
E2
(15%)
E3
(25%)
A–C
10 h extraction time
POS
Aa
23.15
±1.03
22.41Ab
±1.02
23.04Aa
±1.25
Othersγ
SOS
Bb
43.48
±1.23
44.81Bb
±1.57
44.98Aa
±1.65
Ba
28.69
±0.52
28.03Cb
±0.69
27.26Cc
±1.48
Aa
4.68
±0.33
4.75Aa
±0.96
4.72Aa
±0.75
POP
POS
Bb
20.99
±1.03
20.42Ba
±1.23
22.13Bc
±0.98
15 h extraction time
Othersγ
SOS
Aa
45.46
±1.19
45.25Aa
±1.84
45.57Aa
±1.25
Aa
29.42
±0.70
29.89Ba
±1.13
27.86Ba
±1.57
Bb
4.14
±0.56
4.44Ba
±1.99
4.43Ba
±1.01
POP
POS
Cb
20.19
±1.20
20.27Bb
±1.12
21.69Ca
±0.58
Othersγ
SOS
Aa
45.85
±1.26
45.37Aa
±2.36
45.69Aa
±1.05
Ab
29.93
±0.41
30.60Aa
±0.79
28.53Ac
±0.72
4.03Ca
±0.44
3.76Cb
±1.06
4.09Ca
±0.35
Means within a row with different letters are significantly different (p b 0.05).
Means within a column with different letters are significantly different (p b 0.05).
Means value ± standard deviation of three replications.
β
Cocoa butter extracted obtained from cocoa liquor by ethanol as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min with concentration of ethanol 5% (E1), ethanol 15% (E2)
and ethanol 25% (E3).
γ
PLiO, PLiP, POO, SOO, and SOA where P = palmitic, O = oleic, S = stearic, Li = linoleic, A = arachidic.
a–d
α
156
E.K. Asep et al. / Innovative Food Science and Emerging Technologies 20 (2013) 152–160
Table 2
Triacylglycerols composition (area %)α changes of cocoa butter extracted from cocoa liquor with isopropanol as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min with different
isopropanol concentration and extraction time.
Sampleβ
Triacylglycerols compositions
5 h extraction time
POP
POS
Aa
I1
(5%)
I2
(15%)
I3
(25%)
10 h extraction time
21.71
±1.07
21.07Ab
±1.06
21.15Bb
±1.30
SOS
Aa
45.05
±1.19
45.60Aa
±1.52
45.71Aa
±1.59
Others
Ba
29.01
±0.50
29.12Ca
±0.67
28.84Aa
±1.43
γ
Cab
4.23
±0.26
4.21Cb
±0.75
4.30Ca
±0.59
POP
15 h extraction time
POS
Bc
SOS
Bb
20.39
±0.91
21.04Ab
±1.09
22.03Aa
±0.87
44.05
±0.90
42.74Bc
±1.40
45.83Aa
±0.96
Others
Ab
30.57
±0.51
31.18Ba
±0.81
27.73Bc
±1.13
γ
POP
Aa
POS
Bc
3.99
±0.33
5.03Aa
±1.16
4.41Bb
±0.59
20.14
±1.92
20.58Bb
±1.80
22.00Aa
±0.93
Othersγ
SOS
Bb
40.07
±1.01
42.59Bc
±1.89
45.77Aa
±0.84
Ab
30.92
±0.62
31.85Aa
±1.19
27.70Bc
±1.08
4.87Bb
±0.34
4.98Bb
±0.82
4.53Ac
±0.27
A–C
Means within a row with different letters are significantly different (p b 0.05).
Means within a column with different letters are significantly different (p b 0.05).
Means value ± standard deviation of three replications.
β
Cocoa butter extracted obtained from cocoa liquor by ethanol as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min with concentration of isopropanol 5% (I1), isopropanol 15%
(I2) and isopropanol 25% (I3).
γ
LiO, PLiP, POO, SOO, and SOA where P = palmitic, O = oleic, S = stearic, Li = linoleic, A = arachidic.
a–d
α
Table 3
Triacylglycerols composition (area %)α changes of cocoa butter extracted from cocoa liquor with acetone as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min with different acetone concentration and extraction time.
Sampleβ
Triacylglycerols compositions
5 h extraction time
POP
Aa
AC1
(5%)
AC2
(15%)
AC3
(25%)
10 h extraction time
POS
23.01
±1.16
22.37Ab
±1.15
21.40Bc
±1.41
Othersγ
SOS
Bb
44.33
±1.12
45.30Aab
±1.43
44.82Ba
±1.50
Cb
28.51
±0.50
27.98Cc
±0.66
29.11Aa
±1.41
Ac
4.14
±0.27
4.35Bb
±0.81
4.67Aa
±0.63
POP
15 h extraction time
POS
Bb
Aa
21.11
±0.94
21.14Bb
±1.12
22.03Aa
±0.90
Othersγ
SOS
45.15
±1.03
42.94Bb
±1.60
45.83Aa
±1.09
Bb
29.59
±0.69
31.33Ba
±1.10
27.73Bc
±1.54
POP
Ac
POS
Cc
4.16
±0.42
4.59Aa
±1.49
4.41Bb
±0.76
20.34
±1.52
21.07Bb
±1.42
21.59Ba
±0.74
Othersγ
SOS
Aa
45.44
±1.01
43.21Bb
±1.89
45.95Aa
±0.84
Ab
30.16
±0.53
31.81Aa
±1.03
28.39Cc
±0.93
4.06Ba
±0.39
3.91Cb
±0.95
4.07Ca
±0.31
A–C
Means within a row with different letters are significantly different (p b 0.05).
Means within a column with different letters are significantly different (p b 0.05).
α
Means value ± standard deviation of three replications.
β
Cocoa butter extracted obtained obtained from cocoa liquor by acetone as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min with concentration of acetone 5% (AC1), acetone
15% (AC2) and acetone 25% (AC3).
γ
PLiO, PLiP, POO, SOO, and SOA where P = palmitic, O = oleic, S = stearic, Li = linoleic, A = arachidic.
a–d
1.40
5% EtOH - POP/POS
15% EtOH - POP/POS
25% EtOH - POP/POS
5% EtOH - SOS/POS
15% EtOH - SOS/POS
25% EtOH - SOS/POS
5% EtOH - Others/POS
15% EtOH - Others/POS
25% EtOH - Others/POS
1.30
1.20
1.40
5% Isopropanol - POP/POS
15% Isopropanol - POP/POS
25% Isopropanol - POP/POS
5% Isopropanol - SOS/POS
15% Isopropanol - SOS/POS
25% Isopropanol - SOS/POS
5% Isopropanol - Others/POS
15% Isopropanol - Others/POS
25% Isopropanol - Others/POS
1.30
1.10
1.20
Selectivity (α)
Selectivity (α).
The results also show that POS, SOS and POP were the major TGs
extracted in decreasing order (Tables 1, 2 and 3); for ethanol, POS
(43.48–45.85%), SOS (27.26–30.60%) and POP (20.19–23.15%); for
isopropanol POS (40.07–45.83%), SOS (27.70–31.85%) and POP (20.14–
1.00
0.90
0.80
1.10
1.00
0.90
0.70
0.80
0
5
10
15
20
Extraction Time (h)
Fig. 4. Selectivity (α) of triacylglycerols (TG) as function of ethanol concentration as
cosolvent and extraction time on cocoa butter extraction using SC–CO2 at 35 MPa
and 60 °C, whereas others TG are PLiO, PLiP, POO, SOO and SOA.
0
5
10
15
20
Extraction Time (h)
Fig. 5. Selectivity (α) of triacylglycerols (TG) as function of isopropanol concentration
as cosolvent and extraction time on cocoa butter extraction using SC–CO2 at 35 MPa
and 60 °C, whereas others TG are PLiO, PLiP, POO, SOO and SOA.
E.K. Asep et al. / Innovative Food Science and Emerging Technologies 20 (2013) 152–160
1.40
5% Acetone - POP/POS
15% Acetone - POP/POS
25% Acetone - POP/POS
5% Acetone - SOS/POS
15% Acetone - SOS/POS
25% Acetone - SOS/POS
5% Acetone - Others/POS
15% Acetone - Others/POS
25% Acetone - Others/POS
1.30
Selectivity (α)
.
1.20
1.10
1.00
0.90
0.80
0.70
0
5
10
15
20
Extraction Time (h)
Fig. 6. Selectivity (α) of triacylglycerols (TG) as function of acetone concentration as
cosolvent and extraction time on cocoa butter extraction using SC–CO2 at 35 MPa
and 60 °C, whereas others TG are PLiO, PLiP, POO, SOO and SOA.
22.03%); and for acetone, POS (42.94–45.95%), SOS (27.73–31.81%) and
POP (20.34–23.01%). The proposed TG profiles of cocoa butter were similar to the typical TG composition reported by Rossi et al. (1989).
157
The results also clearly indicated that the TG profile was significantly (p b 0.05) influenced by the cosolvent concentration and extraction time (Tables 1, 2 and 3). For instance, the percentage of
POP decreased with an increase in the concentration of cosolvent
and extraction time. Conversely, the percentages of POS and SOS increased as the cosolvent concentration and extraction time were increased. These observations could be explained by the fact that POP
was more soluble than POS and SOS. Hence, it was differentiated by
the first stage of SC–CO2 extraction process, while POS and SOS
were eluted in the second stage of SC–CO2 extraction.
The separation of TGs during SC–CO2 extraction was clearly shown
by the selectivity (α) in Figs. 4, 5 and 6, where the selectivity (α) variation of TGs could be seen as a function of polar cosolvent concentration and extraction time. The selectivity of POP increased with the
addition of 5% ethanol, followed by 15% and 25% ethanol (Fig. 4).
However, the concentration of ethanol had significant (p b 0.05) negative effects on the selectivity of SOS, in which the selectivity of SOS
decreased with increasing concentration of ethanol. The highest
selectivity of SOS was obtained by using 5% ethanol (Fig. 4). Furthermore, the selectivity of POP inconsistently changed with an increase
in the isopropanol concentration. A significant change (p b 0.05) in
the selectivity of POP and SOS was achieved by the addition of
5% and 15% isopropanol, respectively, while the addition of 25%
isopropanol did not produce a significant change (p N 0.05) in the selectivity of SOS (Fig. 5). Similarly, the selectivity of POP was inconsistently affected as the acetone concentration was increased (Fig. 6).
The addition of 5% acetone produced a significant change (p b 0.05)
in the selectivity of POP, whereas that of 15% or 25% isopropanol did
cause a significant change (p N 0.05) for POP.
The selectivity of TG was significantly (p b 0.05) affected by extraction time and cosolvent concentration. This finding could be due
Table 4
Fatty acid composition (area %)α changes of cocoa butter extracted from cocoa liquor by ethanol as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min of flow rate with different
ethanol concentration and extraction time.
Sampleβ
5 h extraction time
C16:0
E1
(5%)
E2
(15%)
E3
(25%)
C18:0
Aa
29.90
±1.43
29.35Ab
±1.41
29.09Ab
±1.74
10 h extraction time
C18:1
Aa
34.95
±1.36
35.04Ba
±1.74
35.58Ba
±1.83
Othersγ
C18:2
Aa
31.44
±0.46
31.72Aa
±0.61
31.33Aa
±1.31
Bb
2.38
±0.56
2.51Ba
±1.64
2.52Aa
±1.28
Ac
1.33
±0.66
1.68Aa
±1.88
1.48Ab
±1.46
C16:0
C18:0
Ba
29.00
±1.41
29.06Aa
±1.33
28.37Bb
±1.55
15 h extraction time
C18:1
Ab
35.57
±1.04
35.13Bb
±1.26
36.95Aa
±1.26
Othersγ
C18:2
Aa
31.82
±0.32
31.72Aa
±0.41
31.01Ab
±0.83
Ab
2.46
±0.36
2.60Aa
±1.00
2.36Bc
±0.74
Bc
1.16
±0.45
1.50Ca
±1.22
1.31Bb
±0.91
C16:0
C18:0
Ca
28.03
±1.75
28.27Ba
±1.66
28.03Ba
±1.08
C18:1
Aa
36.93
±0.34
36.52Aa
±0.81
37.02Aa
±1.08
Othersγ
C18:2
Aa
31.61
±0.98
31.24Aa
±2.13
31.37Aa
±1.03
Ba
2.38
±0.77
2.39Ca
±1.83
2.31Cb
±0.88
1.05Cc
±0.50
1.57Ba
±0.57
1.27Cb
±0.24
A–C
Means within a row with different letters are significantly different (p b 0.05).
Means within a column with different letters are significantly different (p b 0.05).
Means value ± standard deviation of three replications.
β
Cocoa butter extracted from cocoa liquor obtained by ethanol as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min with concentration of ethanol 5% (E1) ethanol 15% (E2) and
ethanol 25% (E3).
γ
C12:0, C14:0 and C18:3.
a–e
α
Table 5
Fatty acid composition (area %)α changes of cocoa butter extracted from cocoa liquor by isopropanol as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min of flow rate with different isopropanol concentration and extraction time.
Sampleβ
5 h extraction time
C16:0
I1
(5%)
I2
(15%)
I3
(25%)
A–C
C18:0
Aa
29.60
±0.68
28.32Ab
±0.98
28.51Ab
±0.97
10 h extraction time
C18:1
Aa
35.07
±1.11
36.40Ba
±1.36
37.90Ba
±1.74
Othersγ
C18:2
Aa
31.53
±0.61
31.05Aa
±0.45
30.14Aa
±0.60
Bb
2.56
±0.95
2.26Ba
±0.51
2.58Aa
±1.50
Ac
1.24
±0.56
1.97Aa
±0.53
0.87Ab
±1.50
C16:0
C18:0
Ba
29.15
±0.97
29.81Aa
±1.34
28.03Bb
±1.26
15 h extraction time
C18:1
Ab
37.30
±0.87
35.70Bb
±1.02
36.92Aa
±1.24
Othersγ
C18:2
Aa
30.03
±0.45
30.54Aa
±0.32
30.34Ab
±0.40
Ab
2.35
±0.62
2.69Aa
±0.32
2.64Bc
±0.89
Bc
1.17
±0.27
1.27Ca
±0.24
1.08Bb
±0.64
C16:0
C18:0
Ca
29.23
±1.43
29.59Ba
±1.70
28.37Ba
±1.61
C18:1
Aa
37.20
±0.30
35.81Aa
±0.24
38.38Aa
±0.58
Othersγ
C18:2
Aa
30.10
±1.55
30.64Aa
±0.89
29.74Aa
±1.94
Ba
2.29
±0.60
2.70Ca
±0.60
2.43Cb
±1.42
1.18Cc
±0.16
1.26Ba
±0.21
1.09Cb
±0.24
Means within a row with different letters are significantly different (p b 0.05).
Means within a column with different letters are significantly different (p b 0.05).
Means value ± standard deviation of three replications.
β
Cocoa butter extracted obtained from cocoa liquor by isopropanol as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min with concentration of isopropanol 5% (I1) isopropanol
15% (I2) and isopropanol 25% (I3).
γ
C12:0, C14:0 and C18:3.
a–e
α
158
E.K. Asep et al. / Innovative Food Science and Emerging Technologies 20 (2013) 152–160
Table 6
Fatty acid composition (area %)α changes of cocoa butter extracted from cocoa liquor by acetone as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min of flow rate with different
acetone concentration and extraction time.
Sampleβ
5 h extraction time
C16:0
AC1
(5%)
AC2
(15%)
AC3
(25%)
C18:0
Aa
29.79
±1.03
29.60Ab
±1.02
29.36Ab
±1.25
10 h extraction time
C18:1
Aa
36.50
±1.01
34.65Ba
±1.29
34.61Ba
±1.36
Othersγ
C18:2
Aa
29.98
±0.31
31.98Aa
±0.41
31.12Aa
±0.87
Bb
2.49
±0.56
2.53Ba
±1.65
2.57Aa
±1.29
Ac
1.24
±0.59
1.37Aa
±1.69
1.34Ab
±1.32
C16:0
C18:0
Ba
29.81
±1.40
29.33Aa
±1.32
30.38Bb
±1.54
15 h extraction time
C18:1
Ab
36.57
±1.02
35.36Bb
±1.24
35.21Aa
±1.24
Othersγ
C18:2
Aa
30.00
±0.47
30.83Aa
±0.59
30.59Ab
±1.20
Ab
2.53
±0.48
2.62Aa
±1.34
2.53Bc
±0.99
C16:0
Bc
C18:0
Ca
1.09
±0.36
1.86Ca
±0.98
1.53Bb
±0.72
29.54
±1.31
28.89Ba
±1.25
28.71Ba
±0.81
C18:1
Aa
37.14
±0.30
36.26Aa
±0.73
37.92Aa
±0.97
Othersγ
C18:2
Aa
29.73
±0.60
30.87Aa
±1.31
29.71Aa
±0.63
Ba
2.51
±0.39
2.45Ca
±0.92
2.36Cb
±0.44
1.08Cc
±0.35
1.53Ba
±0.41
1.29Cb
±0.17
A–C
Means within a row with different letters are significantly different (p b 0.05).
Means within a column with different letters are significantly different (p b 0.05).
Means value ± standard deviation of three replications.
β
Cocoa butter extracted obtained from cocoa liquor by acetone as cosolvent in SC–CO2 at 35 MPa, 60 °C and 2 mL/min with concentration of acetone 5% (AC1) acetone 15% (AC2)
and acetone 25% (AC3).
γ
C12:0, C14:0 and C18:3.
a–e
α
Furthermore, the results show that C16:0 (34.61–37.92%), followed
by C18:1 (29.71–31.98%) and C18:0 (28.71–30.38%), was found to
be the major fatty acid, as acetone was added as a polar cosolvent.
As shown in Tables 4, 5 and 6, C16:0 and C18:0 profiles decreased as
the cosolvent concentration and the extraction time were increased. On
the other hand, C18:1 and C18:2 increased with an increase in the
cosolvent concentration and prolonged extraction time. These results
indicate that during extraction, more soluble fatty acids, C16:0 and
C18:0, came out in the first stage of extraction using SC–CO2, followed
by less soluble fatty acids C18:1, and C18:2, which came out in the second stage.
a) C16:0 and C18:1
1.70
5% EtOH - C16:0/C18:0
15% EtOH - C16:0/C18:0
25% EtOH - C16:0/C18:0
5% EtOH - C18:1/C18:0
15% EtOH - C18:1/C18:0
25% EtOH - C18:1/C18:0
1.60
1.50
.
Selectivity (α)
to the difference between the solubility of the three main TGs (i.e.
POP, SOS and POS) (Asep et al., 2008; Scheider et al., 2000; Soares,
Gamarra, Paviani, Gonçalves, & Cabral, 2007). In this case, the amount
of short-chain TGs decreased, whereas that of the long-chain TGs increased. This observation was also reported in a previous study (Arul,
Boudreau, Makhlouf, Tardif, & Sahasrabudhe, 1987). The results
indicate that the variation of POP composition was significantly
(p b 0.05) higher when the extraction process was performed using
5% ethanol (Fig. 4), followed by 15% isopropanol (Fig. 5) and 5% acetone (Fig. 6). In fact, the addition of a low concentration of polar
cosolvents resulted in the highest selectivity and excessive concentration resulted in less selectivivity for POP.
The same trend was observed for SOS. The highest variation of
SOS selectivity was obtained by using 15% ethanol, followed by 5%
and 25% ethanol (Fig. 4). However, the addition of 5% or 25% polar
cosolvent did not significantly (p N 0.05) induce a change in the SOS
selectivity. Opposite patterns in the selectivity of POP and SOS were
observed. In fact, POP was more quickly extracted at the beginning
of the extraction process whereas SOS was more quickly extracted
at the end. Soares et al. (2007) reported that the solubility of the
pure TGs was significantly (p b 0.05) influenced by the temperature
and length of fatty acid chains. The authors also found that the solubility of the supercritical carbon dioxide varied as a function of the
molecular weight of triacylglycerols containing saturated fatty acids.
In fact, the low-molecular-weight triacylglycerols provided higher
solublility compared to the high molecular weight triacylglycerols.
This observation indicates that lower-molecular-weight POP would
be extracted more quickly than the higher-molecular-weight SOS.
1.40
1.30
1.20
1.10
1.00
0.90
0
5
3.3. Effects on fatty acid methyl ester profile and selectivity (α)
15
20
b) C18:2 and others FA (C12:0, C14:0, C18:3 and C20:0)
Selectivity (α)
The results obtained from the GC analysis of fatty acid composition
are shown in Tables 4, 5 and 6. The results showed that C16:0, C18:0
and C18:1 were found to be the three main fatty acids (FA) in the
cocoa butter. In general, the concentration of ethanol and acetone
were varied and showed significant (p N 0.05) effects on the fatty
acid profile, except in the effect of ethanol (E) on C18:0 (E2, E3),
other FA (15 h), and in the effect of acetone (AC) on C16:0 (AC3),
C18:0 (AC2, AC3), C16 (5, 10, 15 h), C18:0 (5, 10, 15 h), C18:1 (5,
10 h) and other FA (10, 15 h). On the other hand, the concentration
of isopropanol was also varied and had a significant (p b 0.05) effect
on the fatty acid profile, except for C16:0 (I3), C18:0 (I3), C18:1 (I1,
I3), C18:2 (I1, I3), other FA (I1, I3), C18:1 (15 h), C18:2 (5, 10, 15 h)
and other FA (10 h). As shown in Tables 4 to 6, C18:0 (34.95–
37.02%) followed by C18:1 (31.01–31.82%) and C16:0 (28.03–
29.90%) were found to be the three major fatty acids as ethanol was
used as a polar cosolvent. The results also show that C18:0 (35.07–
38.38%), C18:1 (30.03–31.53%) and C16:0 (28.03–29.59%) were
identified as the major FA when isopropanol was used as a cosolvent.
10
Extraction Time (h)
1.80
1.70
1.60
1.50
1.40
1.30
1.20
1.10
1.00
0.90
0.80
5% EtOH - C18:2/C18:0
15% EtOH - C18:2/C18:0
25% EtOH - C18:2/C18:0
5% EtOH - Others/C18:0
15% EtOH - Others/C18:0
25% EtOH - Others/C18:0
0
5
10
15
20
Extraction Time (h)
Fig. 7. Selectivity (α) of fatty acid methyl esters as function of ethanol concentration as
cosolvent and extraction time on cocoa butter extraction using SC–CO2 at 35 MPa and
60 °C.
E.K. Asep et al. / Innovative Food Science and Emerging Technologies 20 (2013) 152–160
a) C16:0 and C18:1
1.70
5% Isopropanol - C16:0/C18:0
15% Isopropanol - C16:0/C18:0
25% Isopropanol - C16:0/C18:0
5% Isopropanol - C18:1/C18:0
15% Isopropanol - C18:1/C18:0
25% Isopropanol - C18:1/C18:0
5% Acetone - C16:0/C18:0
15% Acetone - C16:0/C18:0
25% Acetone - C16:0/C18:0
5% Acetone - C18:1/C18:0
15% Acetone - C18:1/C18:0
25% Acetone - C18:1/C18:0
1.60
1.50
Selectivity (α)
.
Selectivity (α)
a) C16:0 and C18:1
1.80
1.70
1.60
1.50
1.40
1.30
1.20
1.10
1.00
0.90
0.80
1.40
1.30
1.20
1.10
1.00
0.90
0
5
10
15
20
0.80
Extraction Time (h)
0
5
5% Isopropanol - C18:2/C18:0
15% Isopropanol - C18:2/C18:0
25% Isopropanol - C18:2/C18:0
5% Isopropanol - Others/C18:0
15% Isopropanol - Others/C18:0
25% Isopropanol - Others/C18:0
.
1.60
2.20
Selectivity (α)
1.20
1.00
0.80
0.60
5
10
15
20
5% Acetone - C18:2/C18:0
15% Acetone - C18:2/C18:0
25% Acetone - C18:2/C18:0
5% Acetone - Others/C18:0
15% Acetone - Others/C18:0
25% Acetone - Others/C18:0
2.00
1.40
0
15
b) C18:2 and others FA (C12:0, C14:0, C18:3 and C20:0)
2.00
1.80
10
Extraction Time (h)
b) C18:2 and others FA (C12:0, C14:0, C18:3 and C20:0)
Selectivity (α)
159
20
Extraction Time (h)
Fig. 8. Selectivity (α) of fatty acid methyl esters as function of isopropanol concentration as cosolvent and extraction time on cocoa butter extraction using SC–CO2 at
35 MPa and 60 °C.
Our study clearly shows that in the extraction of cocoa butter using
SC–CO2 at 35 MPa 60 °C, the polar cosolvent's concentration and extraction time show significant effects (p b 0.05) on the selectivity (α).
In general a 25% polar cosolvent concentration showed the largest increase in C16:0 selectivity (α), followed by 15 and 5%, respectively.
The effects of ethanol, isopropanol and acetone concentrations on the
selectivity for different FAs are shown in Figs. 7, 8 and 9, respectively.
There was a large increase in the selectivity of C16:0 with an increase
in extraction time and the addition of 25%, followed by 15% ethanol.
However, C18:1, C18:2 and other FAs showed a small variation in the
selectivity with the addition of ethanol for all cosolvent concentrations.
It is clear that C16:0 was separated more rapidly compared to C18:1,
C18:2 and other FAs during extraction (Fig. 7a–b). For isopraponol, a
larger increase in the selectivity of C16:0 was observed with the addition of 25% ethanol, followed by 15% ethanol. Similarly, other FAs
showed large changes in the selectivity with the addition of 25%
isopropanol (Fig. 8). However, unsaturated FAs (C18:1, C18:2) showed
a small change in selectivity. As with ethanol, the saturated C16:0 was
also separated more rapidly compared to the unsaturated fatty acids,
C18:1 and C18:2, during the SC–CO2 extraction of cocoa butter using
isopropanol. These results are consistent with those of other studies
on SC–CO2 fractionation of crude palm oil (Markom, Singh, & Hasan,
2001; Zaidul, Norulaini, Mohd Omar, & Smith, 2007a, 2007b), in
which the lower-molecular-weight compounds such as saturated FAs
were separated first, followed by the higher-MW compounds such
as unsaturated FAs. The same effect of acetone concentration on the
selectivity of fatty acids (Fig. 9) also showed the same trend when
isopropanol was used as a cosolvent.
1.80
1.60
1.40
1.20
1.00
0.80
0
5
10
15
20
Extraction Time (h)
Fig. 9. Selectivity (α) of fatty acid methyl esters as function of acetone concentration as
cosolvent and extraction time on cocoa butter extraction using SC–CO2 at 35 MPa and
60 °C.
4. Conclusion
The cocoa butter was successfully extracted from cocoa liquor by
SC–CO2 at 35 MPa, 60 °C and 2 mL/min using different concentrations
of polar cosolvents (ethanol, isopropanol and acetone). The extraction
yield was significantly (p b 0.05) influenced by the concentration of
polar cosolvents. Similarly, polar cosolvent concentration had significant (p b 0.05) effects on the TG and FA compositions. Ethanol was
found to be the most efficient polar cosolvent for cocoa butter extraction compared to isopropanol and acetone. POS (42.2–45.9%) being
the major triglycerdies component, followed by SOS (27.6–31.4%)
and POP (20.3–22.7). Palmitic, stearic and oleic acids were the main
fatty acids in the extracted cocoa butter, with stearic being the highest
(34.9-37.8%), followed by oleic (30.3-31.8%) and palmitic (28.3-30.0%)
acids, respectively. In terms of the selectivity, the lower-molecularweight of TGs and FAs showed higher selectivity compared to
the higher-molecular-weight TGs and FAs; therefore, POP was the
major triglyceride at the beginning of extraction, while POS, followed
by SOS, were the major triglycerides at the end of the extraction process. However, the FA composition was found to be more selective
when a high concentration of polar cosolvent (25%) was used. The
choice of modifiers becomes a great challenge and ethanol was shown
to be the best polar cosolvent, and it enhanced the solubility during
the cocoa butter extraction by SC–CO2. This method can be feasibly
implemented in the cocoa industry for the production of high quality
cocoa butter.
160
E.K. Asep et al. / Innovative Food Science and Emerging Technologies 20 (2013) 152–160
Acknowledgements
We thank The Ministry of Science, Technology and Environment of
Malaysia which has sponsored this research under Intensification of
Research Priority Area (IRPA) Project No. 01-02-04-0512.
References
American Oil Chemist Society (AOCS) (1993). Official and Tentative Methods of American
Oil Chemist Society (4th ed.) (pp. 1–5). Champaign, Illinois, USA: American Oil
Chemist's Society (AOCS).
Arul, J., Boudreau, A., Makhlouf, J., Tardif, R., & Sahasrabudhe, M. R. (1987). Fractionation of
anhydrous milk fat by supercritical carbon dioxide. Journal of Food Science, 52,
1231–1235.
Asep, E. K., Jinap, S., Russly, A. R., Harcharan, S., & Nazimah, S. A. H. (2008). The effects
of particle size, fermentation and roasting of cocoa nibs on supercritical fluid extraction of cocoa butter. Journal of Food Engineering, 85, 450–458.
Association of Official Analytical Chemists (AOAC) (1998). Official methods of analysis
(17th ed.). Gaithersburg, MD, USA: Association of Official Analytical Chemists
(AOAC).
Catchpole, O. J., Grey, J. B., & Noermark, K. A. (1998). Solubility of fish oil components in
supercritical CO2 and CO2 + ethanol mixtures. Journal of Chemical & Engineering
Data, 43, 1091–1095.
Cocero, M. J., & Calvo, L. (1996). Supercritical fluid extraction of sunflower seed oil with
CO2-ethanol mixtures. Journal of the American Oil Chemists' Society, 73, 1573–1574.
Dauksas, E., Venskutonis, P. R., & Sivik, B. (1998). Extraction of lovage (Levisticum
officinale Koch.) roots by carbon dioxide. 1. Effect of CO2 parameters on the yield
of the extract. Journal of Agricultural and Food Chemistry, 46, 4347–4352.
Dobbs, J. M., Wong, J. M., Lahiere, R. J., & Johnston, K. P. (1987). Modification of supercritical fluid phase behavior using polar co-solvents. Industrial and Engineering
Chemistry Research, 26, 56–62.
Francisco, J. D. C., & Dey, E. S. (2003). Supercritical fluids as alternative, safe, foodprocessing media: an overview. Acta Microbiologica Polonica, 52, 35–39.
Hawthorne, S. B. (1990). Analytical Scale Supercritical Fluid Extraction. Journal of
Analytical Chemistry, 62, 633–639.
King, J. W. (1993a). Analysis of fats and oils by SFE and SFC. INFORMA, 4, 1089–1096.
King, J. W. (1993b). Hazardous waste reduction in the development of analytical
methods—Supercritical fluid extraction. Symposium on new and emerging approaches
to the development of methods in trace-level analytical chemistry, Springs, Berkeley.
King, J. W., & France, J. E. (1992). Basic principles of analytical supercritical fluid extraction.
In B. Wenclawiak (Ed.), Analysis with supercritical fluids. Berlin: Springer-Verlag.
Li, S., & Hartland, S. (1992). Influence of co-solvents on solubility and selectivity in extraction of xanthines and cocoa butter from cocoa beans with supercritical CO2.
Journal of Supercritical Fluids, 5, 7–12.
Li, S., & Hartland, S. (1996). A new industrial process for extracting cocoa butter and
xanthines with supercritical carbon dioxide. Journal of the American Oil Chemists'
Society, 73, 423–429.
Luque de Castro, M. D., & Garcia-Ayuso, L. E. (1998). Soxhlet extraction of solid materials:
An outdated technique with a promising innovative future. Analytica Chemica Acta,
369(1–2), 1–10.
Machmudah, S., Shotipruk, A., Goto, M., Sasaki, M., & Hirose, T. (2006). Extraction of
astaxanthin from Haematococcus pluvialis using supercritical CO2 and ethanol as
entrainer. Industrial and Engineering Chemistry Research, 45, 3652–3657.
Markom, M., Singh, H., & Hasan, M. (2001). Supercritical CO2 fractionation of crude
palm oil. Journal of Supercritical Fluids, 20, 45–53.
Olimpio, M., Maria Dolores, M., Carmen, M. L., Luis, M. L., Casimiro, M., Miguel, R., et al.
(2005). Supercritical CO2 extraction of b-carotene from a marine strain of the cyanobacterium Synechococcus species. Journal of Agricultural and Food Chemistry, 53,
9701–9707.
Reverchon, E., & Camillis, M. D. (1991). Manufacture of food extracts by SCFE: Rosemary essential oil. Industrie Alimentari, 30, 266–267.
Rossi, M. (1996). Supercritical fluid extraction of cocoa and cocoa products. In J. W.
King, & G. R. List (Eds.), Supercritical fluid technology in oil and lipid chemistry
(pp. 220–229). Champaign, Illinois: AOCS Press.
Rossi, M., Arnoldi, C., Salvioni, G., & Schiraldi, A. (1989). Characterization of cocoa extracts obtained with supercritical carbon dioxide. Italian Journal of Food Science, 3,
41–50.
Scheider, G. M., Kautz, C. B., & Tuma, D. (2000). Supercritical fluids: Fundamental and
applications. In E. Kiran (Ed.), Physico-chemical principles of supercritical fluid science (pp. 31–68). Netherlands: Kluwer Academic Publisher.
Shi, J., Yi, C., Xue, S. J., Jiang, Y., Ma, Y., & Li, D. (2009). Effects of modifiers on the profile of
lycopene extracted from tomato skins by supercritical CO2. Journal of Food Engineering,
93, 431–436.
Soares, B. M. C., Gamarra, F. M. C., Paviani, L. C., Gonçalves, L. A. G., & Cabral, F. A. (2007).
Solubility of triacylglycerols in supercritical carbon dioxide. Journal of Supercritical
Fluids, 43, 25–31.
Stahl, E., Quirin, K. W., & Gerald, D. (1988). Dense gases for extraction and refining. New
York: Springer-Verlag, 176.
Temelli, F. J. (1992). Extraction of triglycerides and phospholipids from canola with supercritical carbon dioxide and ethanol. Journal of Food Science, 57, 440–443.
Walsh, J. M., Greenfield, M. L., Ikonomou, G. D., & Donohue, M. D. (1989). Hydrogen
bonding competition in entrainer cosolvent systems. Chemical Engineering Communications, 86, 125–144.
Walsh, J. M., & Ikonomou, M. D. (1987). Donohue, Supercritical phase behavior: The entrainer effect. Fluid Phase Equilibria, 33, 295–314.
Zaidul, I. S. M., Norulaini, N. N. A., Mohd Omar, A. K., & Smith, R. L., Jr. (2007a). Blending of
supercritical carbon dioxide (SC–CO2) extracted palm kernel oil fractions and palm oil
to obtain cocoa butter replacers. Journal of Food Engineering, 78, 1397–1409.
Zaidul, I. S. M., Norulaini, N. N. A., Mohd Omar, A. K., & Smith, R. L., Jr. (2007b). Supercritical carbon dioxide (SC–CO2) extraction of palm kernel oil from palm kernel.
Journal of Food Engineering, 79, 1007–1014.